Genomic Fingerprinting of Borrelia burgdorfeni Sensu Lato by Pulsed-Field Gel Electrophoresis

JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 993, p. 873-877 Vol. 3, No. 0095-37/93/873-05$0.00/0 Copyright 993, American Society for Microbiology Genomic Fingerprinting of Borrelia burgdorfeni Sensu Lato by Pulsed-Field Gel Electrophoresis J. BELFAIZA, D. POSTIC, E. BELLENGER, G. BARANTON, A I. SAINT GIRONS* Unites de Bact6riologie Moleculaire et Me6dice Institut Pasteur, 5 Rue du docteur Roux, 7574 Paris Cedex 5, Received 4 May 993/Returned for modification 5 July 993/Accepted 7 August 993 A tot of 46 Borrelia burgdorferi sensu lato isolates that were isolated from patients with Lyme borreliosis and infected anims or were extracted from ticks of the genus lxodes were anyzed. Large restriction fragment patterns obtained after cleavage of genomic DNAs with Mlul were anyzed by pulsed-field gel electrophoresis (PFGE). To eliminate the contribution of plasmid DNA, only frmments greater than 70 kb were used for the anysis. The results indicated that each of the 4 B. burgdorferi sensu stricto isolates were recognized by a band at 35 kbp, each of the Borrelia garinii isolates by two bands (0 and 80 kbp), and each of the 0 Borrelia afzelii isolates by three bands (460, 30, and 90 kbp). Whereas differences in the PFGE patterns among B. burgdorferi sensu stricto isolates and B. garinii isolates were noted, B. afzelii isolates were l similar. Identification of isolates by PFGE correlates with their belonging to a given species within B. burgdorferi sensu lato. Borrelia burgdorferi was identified in 98 as the agent of Lyme borreliosis () and was recognized as a new species of the genus Borrelia in 984 (6). Three species were recently delineated for B. burgdorferi sensu lato: B. burgdorferi sensu stricto, Borrelia gainii, and Borrelia afzelii (4, 0). This division into three species was based on DNA- DNA hybridizations, rrna gene restriction patterns, and identification by monoclon antibodies, l tools recognized currently by taxonomists (9). The results obtained by different methods of typing isolates such as polymerase chain reaction (, 3), arbitrarily primed polymerase chain reaction (30), multilocus enzyme electrophoresis (0), 6S rrna sequencing (,, 8, 9, 5), fla gene sequencing (, 8), and serotyping (3) are consistent with this new taxonomy. Heterogeneity among isolates has so been observed by anysis of the profiles of linear and circular plasmids (6, 7). Digestion of genomic DNAs with restriction endonucleases followed by convention electrophoresis so showed differences among isolates. However, the thousands of bands obtained are difficult to interpret except when Southern anysis is used following hybridization with diverse probes (7, 7, 8). The latter technique yielded groupings of isolates ong the same lines as described elsewhere (4, 0). Pulsed-field gel electrophoresis (PFGE) was used for constructing the physic map of B. burgdorferi (). This genome is unique, since it contains a 945-kbp linear chromosome as well as linear and circular plasmids which vary in size and number among the isolates (7, 9, 3). In the present article, we anyze sever different isolates by PFGE. The results indicate that a given isolate can be readily identified as belonging to a specific species. * Corresponding author. Electronic mail address: isgirons@pasteur.fr. 873 MATERIALS A METHODS Bacteri isolates and media. The B. burgdorferi sensu lato isolates used in this study are listed in Table. The isolates were grown in BSKII medium at 30 C (5). PFGE, digestion of DNA in agarose, and large restriction faments patterns LRFP nomenclature. Previously described procedures were used for the preparation of high-molecularweight genomic DNAs and PFGE (9, ). Separation was achieved with a pulse time ramped from 3 to 40 s for 0 h with a contour-clamped homogeneous electric field-drii apparatus (Bio-Rad laboratories, Richmond, Cif.). Restriction endonucleases MluI (A/CGCGT) and SmaI (CCC/GGG) were purchased from Pharmacia. Genomic DNAs, in low-melting-temperature agarose, were digested with 0 to 40 U of restriction endonuclease for 0 h in 00 PI of the buffer recommended by the supplier. The lambda concatemers used as size markers for Fig. were those used in reference (monomer size, 44.3 kbp) and for Fig. and 3 were those commerciized by Tebu (monomer size, 48.5 kbp). An LRFP is defined as a unique PFGE pattern. We ignored bands lower than 70 kb, since these bands would be the results of uncut or cut plasmids, which could complicate the interpretation (4). The LRFPs obtained were designated ML or SM to identify the restriction endonuclease MluI or SmaI, respectively, with the suffix b, g, or a denoting B. burgdorferi sensu stricto, B. garinii, or B. afzelii, respectively. Different LRFPs of the same species are numbered or, etc. DNA-blot anysis. For blot anysis according to Southern (6), DNA fragments separated by PFGE were submitted to depurination and then transferred to Hybond-N nylon membranes (Amersham Internation, Amersham, England). To prepare the probe, the 35-kbp MluI band from isolate B3 was excised from a low-melting-temperature agarose pulsedfield gel and washed with ml of distilled water for h at room temperature. A reduced volume of 50,ul of the sample was heated at 00 C for 0 min. The tube was placed on ice, and 5 pl was labelled by random priming with digoxigenin- -dutp. The probe thus generated was used in hybridization experiments as described elsewhere (8).

874 BELFAIZA ET AL. J. CLIN. MICROBIOL. B. burgdorferi B3 97 HUM5 IRS HUM784 0006 HUM3336 MEN5 Lake 339 SON88 SON335 Z36 305 B. garnii 0047 VS85 VS468 N34 G5 PBi TN PBr IPer9O T5 VS86 53 TABLE. B. burgdorferi sensu lato isolates used in this study and their characteristicsa Isolate Origin Geographic location ML LRFP Ribotype Serotype Electrotype (ET) B. afzelii VS46 PGau Fl UMO IPer3 PSto DK3 DK8 ACAl ECM ECM8 SmSl Ipl Ji P47a BFox HFox M7 46 L dammini Human CSFb L. ncinus L ncinus Peromyscus leucopus L ncinus Human CSF L. icinus Human CSF L persulcatus L. ncinus Human ECMe Human ECMd Human ECMe Apodemus flavicollis L persulcatus Fox Fox Denmark Denmark China China bl b b3 b3 b4 b5 b6 b6 b7 b7 b8 b8 b9 blo gl gl g3 g4 g4 El/H/Rl/Pl El/H/Rl/Pl E/Hl/Rl/Pl El/H/Rl/Pl El/H/Rl/Pl E3/H3/R3/P E3/H4/R3/P E4/H3/R/P3 E3/H4/R/P4 E3/H4/R/P4 E3/H4/R/P4 E4/H3/R/P3 E3/H4/R3/P5 E5/H3/R3/P6 E3/H4/R3/P E3/H4/R3/P4 E3/H3/R3/P4 E6/H5/R4/P7 E7/H5/R4/ E7//R4/ E7/H5/R4/ E7/H6/R4/ E6/H5/R4/ E7/H5/R4/ E6/H5/R4/ E7/H5/R4/ a Ribotypes are from references 0 and ; they have been designated E, H, R, and P to identify the restriction endonucleases EcoRI, HindIII, EcoRV, and PstI, respectively, that were used, followed by numbers to indicate subtypes. Electrotypes are from reference 0, and serotypes are from reference 3. b CSF, cerebrospin fluid. c, not done. d ACA, acrodermatitis chronica atrophicans. I ECM, erythema chronicum migrans. c 0 66 4 3 7 0 6 3 8 7 5 4 3 3 33 33 6 7 8 9 5 RESULTS Choice of a rare cutting restriction endonuclease yielding an evuable pattern. Of the eight restriction endonucleases used for the construction of the physic map of B. burgdorferi sensu stricto isolate (), we chose MluI to compare the different isolates. This enzyme cuts the DNA in sufficient numbers of fragments to low easily interpretable comparisons. Only fragments greater than 70 kb were used to define an LRFP. Fingerprinting with MluI and recognition of the three species. A tot of 46 B. burgdorferi sensu lato isolates were anyzed by MluI digestion of genomic DNA followed by PFGE. The results showed that with MluI, 5 LRFPs were obtained (Table ). The distribution of LRFPs was as follows: 0 LRFPs for B. burgdorferi sensu stricto (4 isolates), 4 LRFPs for B. garinii ( isolates), and LRFP for B. afzelii (0 isolates) (Table ). Specific bands were found for each species. Anysis of the LRFPs obtained lowed the recog-

VOL. 3, 993 GENOME FINGERPRINTING OF B. BURGDORFERI 875 LRFP TABLE. Sizes of the bands obtained with MluI digestion of DNAs of the B. burgdorferi sensu lato isolates used in this study Isolate(s) Presence of band with the following sizes (kbp): 460 40 400 380 30 80 0 70 60 45 35 0 00 90 80 MLb B3 + + + MLb 97 + + + + MLb3 IRS and HUM5 + + + + MLb4 HUM784 + + + + MLb5 0006 + + + + + MLb6 MEN5 and HUM3336 + + + + MLb7 Lake 339 and SON88 + + + + MLb8 Z36 and SON335 + + + + + MLb9 305 + + + + + MLbO + + + + MLg 0047 and VS85 + + + + ML N34, PBi, G5, VS468, TN, PBr, and IPer9O + + + + MLg3 T5 + + + + MLg4 VS86 and 53 + + + MLa B. afzeliia + + + a All 0 B. afzelii isolates in Table. nition of a specific band at 35 kbp for B. burgdorferi sensu stricto (Fig. A). Two bands at 0 and 80 kbp were specific to B. garinii (Fig., lanes to ), while for B. afzelii, three bands (at 460, 30, and 90 kbp) were characteristic (Fig., lanes 3 and 4; Table ). The specific bands noted above for identification of a species are identified by their size only. The question arises as to whether the bands characteristic of a species are indeed geneticly similar. The 35-kbp MluI band of B. burgdorferi sensu stricto isolate B3 hybridizes to the same 35-kbp band in each of the 4 isolates of that species (Fig. B). A similar experiment showed that the 0-kbp MluI band characteristic of B. garinii is indeed the same for each of the isolates tested (data not shown). The experiments were not performed for B. afzelii because of the unique and unambiguous ML pattern consisting of most of the genome (Table ). Fingerprinting with SmaI. Another rare cutting restriction endonuclease was used to differentiate among isolates showing the same MluI profile. SmaI lowed the differentiation of some isolates from each other. For example, among the B. garinii isolates that were indistinguishable with MluI (Table [LRFP ML]), one isolate (PBr) had a unique LRFP with SmaI that was distinct from those of the other four isolates N34, G5, VS468, and TN) (Fig. 3). On the other hand, isolates T5 and PBr, which had different MluI patterns, shared the same SmaI pattern. Among the 0 B. afzelii isolates which were indistinguishable with MluI, 5 isolates shared the SM pattern, with a characteristic quadruplet (Fig. 3, lane 9) while five isolates had LRFPs which varied from that of SM by one addition or one missing band (LRFPs SMa, SMa3, and SMa4 [Fig. 3, lanes 0 to ]). L 34 56789034L 3 4 5 6 78 9034 9 45.5 97 -ml 35 A FIG.. (A) PFGE of MluI restriction digests of genomic B. burgdorfen sensu stricto DNAs. The lanes contain lambda DNA concatemers (lanes L) and DNAs from isolates B3 (lane ), 97 (lane ), IRS (lane 3), HUM5 (lane 4), HUM784 (lane 5), 0006 (lane 6), MEN5 (lane 7), HUM3336 (lane 8), Lake 339 (lane 9), SON88 (lane 0), Z36 (lane ), SON335 (lane ), 305 (lane 3), and (lane 4). The 35-kbp band characteristic of B. burgdorferi sensu stricto is indicated by an arrowhead on the right. (B) Blot of the PFGE gel hybridized with the 35-kbp band from B. burgdorferi isolate B3 DNA. B

876 BELFAIZA ET AL. J. CLIN. MICROBIOL. 66 - _ 77 FIG.. PFGE of MluI restriction digests of geno DNAs and two representatives of B. afzelii DNA. Themlanes contain contrast, only four patterns (MLgl to MLg4) for B. gannii lambda DNA concatemers (lanes L) and DNAs fron n isolates N34 isolates were obtained, and among them, 7 isolates exhibited (lane ), PBi (lane ), G5 (lane 3), VS468 (lane 4), VtS86 (lane 5) the ML profile. 53 (lane 6), TN (lane 7), PBr (lane 8), IPer90 (lane c 9), 0047 (lane B. afzelii appeared very homogeneous whatever the 0), VS85 (lane ), T5 (lane ), PGau (lane 3), anid VS46 (lane method used. For 4 isolates with the same ML pattern, 4). The 0- and 80-kbp bands characteristic of jb. garinii are only two ribotypes were observed with EcoRI (E6 and E7) indicated on the right by arrows. The arrowheads indicate the three and HindIII (H5 and H6), and a unique ribotype was bands (460, 30, and 90 kbp) characteristic of B. afze?lii. observed with EcoRV (R4). Furthermore, Wilske et. (3) 94 97- L 3456 7890L L division of isolates within a Borrelia species. For example, the 9 B. burgdorferi sensu stricto isolates tested, which had the same ribotype ( [Table ]) with enzymes EcoRI, HindIII, EcoRV, and PstI (4), could be separated -.460 into six LRFPs (Table ). On the other hand, the same MLb3 x 30 pattern was observed for two isolates showing two different 0 ribotypes (Table ). B. burgdorfeii sensu stricto, which seemed homogeneous by ribotyping (4, ) and serotyping (serotype according to reference 3), appeared more 80 heterogeneous with regard to the MluI PFGE patterns 0 (MLbl to MLblO). It should be noted that the type strain B3 is unique in lacking the 00-kbp MluI digestion product. B. garinii was very heterogeneous in rrna gene restriclanesb. gat~in as well as in serotyping (Table ) (3). tion patterns In found that six isolates, now known as B. afzelii, that had been isolated from cutaneous lesions l belonged to serotype (3). DISCUSSION The results of multilocus enzyme electrophoresis (0), Species identification. Anysis of PFGE platterns after another technique used to type isolates, were incorporated MluI digestion permitted the specific identificatiiion of each of in the present study (Table ). No correlation between the three Borrelia species associated with Lym4 e borreliosis. electrotype (obtained by multilocus enzyme electrophoresis) One band at 35 kbp was specific for B. burgclorferi sensu and either LRFP or ribotype could be shown. stricto, whereas two bands (0 and 80 kbp) wer*e found in l A single MluI pattern was obtained for each of the 0 B. B. garinii isolates tested. Three bands (460, 30, and 90 kbp) afzelii isolates tested. However, a quarter of them (ACA, were characteristic of B. afzelii. Strong evidence that the VS46, PGau, Fl, and UMO) had distinct electrotypes bands characteristic of a given species werre the same (ET5, ET6, ET7, ET8, and ET9, respectively) (0), and geneticly strengthened our identification procedure. another quarter (Jl,, IPer3, ECM8, and PSto) Genetic diversity. Species identification by IPFGE corre- had distinct SmaI patterns. Thus, the B. afzelii isolates are sponded exactly to that obtained previously wit] h rrna gene not entirely homogeneous, though the isolates that were restriction patterns (). However, subgroups )btained with tested had the same ML pattern and EcoRV ribotype. PFGE did not ways correspond to those olbtained with Epidemiologic significance. Correlation of PFGE patterns ribotyping. The PFGE method usuly lowed further sub- with the geographic locations of isolation led us to conclude that a high genetic diversity exists within limited geographic areas. For example, seven B. burgdorferi sensu stricto isolates that were isolated from Lxodes pacificus in Cifornia exhibited five different LRFPs (MLb3, -4, -6, -7, and -8). Interestingly, LRFPs MLb3 and MLb8 are common to isolates from the and Europe. Also three B. garinii isolates that were isolated from Lxodes ricinus in exhibited three different LRFPs (MLgl, -, and.._45.5-4). In contrast, a single LRFP (ML) was observed for isolates that were isolated from humans,, and Ixodes persulcatus in,, and (formerly USSR). However, the most striking feature is the monomorphism of the 0 B. afzeldi isolates, regardless of geographic origin or host from which they were isolated (I. ncinus, L pacificus, rodents, foxes, and humans). Recently, PFGE has been used to compare genomic restriction patterns among different bacteri genera. This FIG. 3. PFGE of SmaI restriction digests of genc)mic B. garinii relatively simple technique is applicable to a wide range of and B. afzelii DNAs. The lanes contain lambda DN) A concatemers bacteria. For example, considerable restriction fragment (lanes L), and DNAs from isolates N34 (lane ), G5 ((lane ), VS468 polymorphisms were seen for Leptospira interrogans sensu (lane 3), TN (lane 4), VS85 (lane 5), T5 (lane 6), VS86 (lane 8), DK8 (lane 9), Jl a PBr (lane 7), lato (4, 5). Furthermore, a given NotI fingerprint could be (lane 0), IPer3 (lane ), and PSto attributed to each of the 60 serovars tested (4, 5), (lane ). LRFP SM, exemplified in lane 9 by iso shared by 5 B. afzelii isolates. LRFP SMa, exempl ifiedkinlane 0 lowing the identification of new isolates by LRFP. In by isolate Jl, was shared by isolate. ilrfp SMa3n contrast to the L. interrogans example, restriction anysis exemplified in lane by isolate IPer3, was shared by isolate ECM8. with XbaI could not distinguish among biovars in the genus LRFP SMa4 was exemplified in lane by isolate P' Sto. Brucella (3).

VOL. 3, 993 PFGE anysis of Borrelia genomic DNAs provides a relatively simple means of characterizing Lyme borreliosis spirochetes. Examination of addition isolates would provide further evidence regarding the utility of this approach for taxonomic and epidemiologic purposes. ACKNOWLEDGMENTS This work was supported by grants from the Gould Foundation to the Pasteur Institute. We are grateful to F. Milward, C. Kodner, M. Janda, A. Vogt, J. Anderson, 0. Peter, Ackerman, B. Wilske, V. Preac-Mursic, S. Bergstrom, R. C. Johnson, E. Asbrinck, R. Marconi, and E. Isogai for gifts of isolates. We thank Barrie Davidson and Iain Old for advice and Nyles Charon and Steve Norris for helpful comments on the manuscript. REFERENCES. Adam, T., G. S. Gassmann, C. Rasiah, and U. B. Gobel. 99. Phenotypic and genotypic anysis of Borrelia burgdorferi isolates from various sources. Infect. Immun. 59:579-585.. Adam, T., B. Graf, U. Neubert, and U. B. Gobel. 99. Detection and classification of Borrelia burgdorferi by direct sequencing of 6S rrna amplified after reverse transcription. Med. Microbiol. Lett. :0-6. 3. Allardet-Servent, A., G. Bourg, M. Ramuz, M. Pages, M. Bellis, and G. Roizes. 988. DNA polymorphism in isolates of the genus Brucella. J. Bacteriol. 70:4603-4607. 4. Baranton, G., D. Postic, I. Saint Girons, P. Boerlin, J. C. Piffaretti, M. Assous, and P. A. D. Grimont. 99. Delineation of Borrelia burgdorfen sensu stricto, Borrelia gannii sp. nov., and group VS46 associated with Lyme borreliosis. Int. J. Syst. Bacteriol. 4:378-383. 5. Barbour, A. G. 984. Isolation and cultivation of Lyme disease spirochetes. Ye J. Biol. Med. 57:5-55. 6. Barbour, A. G. 988. Plasmid anysis of Borrelia burgdorfen, the Lyme disease agent. J. Clin. Microbiol. 6:475-478. 7. Barbour, A. G., and C. F. Garon. 987. Linear plasmids of the bacterium Borrelia burgdorfeni have covently closed ends. Science 37:409-4. 8. Baril, C., J. L. Herrmann, C. Richaud, D. Margarita, and I. Saint Girons. 99. Scattering of rrna genes on the physic map of the circular chromosome of Leptospira interrogans serovar icterohaemorrhagiae. J. Bacteriol. 74:7566-757. 9. Banl, C., C. Richaud, G. Baranton, and I. Saint Girons. 989. Linear chromosome of Borrelia burgdorferi. Res. Microbiol. 40:507-56. 0. Boerlin, P., 0. Peter, A. G. Bretz, D. Postic, G. Baranton, and J. C. Piffaretti. 99. Population genetic anysis of Borrelia burgdorferi isolates by multilocus enzyme electrophoresis. Infect. Immun. 60:677-683.. Burgdorfer, W., A. G. Barbour, S. F. Hayes, J. L. Benach, E. Grunwdt, and J. P. Davis. 98. Lyme disease-a tick-borne spirochetosis? Science 6:37-39.. Davidson, B., J. H. MacDougl, and I. Saint Girons. 99. Physic map of the linear chromosome of the bacterium Borrelia burgdorferi, the causative agent of Lyme disease. J. Bacteriol. 74:3766-3774. 3. Ferdows, M. S., and A. G. Barbour. 989. Megabase-sized linear DNA in the bacterium Borrelia burgdorfeni, the Lyme disease agent. Proc. Natl. Acad. Sci. 86:5969-5973. 4. Herimann, J. L., C. Baril, E. Bellenger, P. Perolat, G. Baranton, and I. Saint Girons. 99. Genome conservation in isolates of Leptospira interrogans. J. Bacteriol. 73:758-7588. 5. Herrmann, J. L., E. Bellenger, P. Perolat, G. Baranton, and I. Saint Girons. 99. Pulsed-field electrophoresis of NotI digests of leptospir DNA: a new rapid method of serovar identifica- GENOME FINGERPRINTING OF B. BURGDORFERI 877 tion. J. Clin. Microbiol. 30:696-70. 6. Johnson, R. C., G. P. Schmid, F. W. Hyde, A. G. Steigerwt, and D. J. Brenner. 984. Borrelia burgdorferi sp. nov.: etiologic agent of Lyme disease. Int. J. Syst. Bacteriol. 34:496-497. 7. LeFebvre, R. B., G. C. Perng, and R. C. Johnson. 989. Characterization of Borrelia burgdorfen isolates by restriction endonuclease anysis and DNA hybridization. J. Clin. Microbiol. 7:636-639. 8. Marconi, R. T., and C. F. Garon. 99. Phylogenetic anysis of the genus Borrelia: a comparison of North American and European isolates of Borrelia burgdorfen. J. Bacteriol. 74:4-44. 9. Marconi, R. T., and C. F. Garon. 99. Identification of a third genomic group of Borrelia burgdorferi through signature nucleotide anysis and 6S rrna sequence determination. J. Gen. Microbiol. 38:533-536. 0. Marin Canica, M., F. Nato, L. du Merle, J. C. Mazie, G. Baranton, and D. Postic. Monoclon antibodies for identification of Borrelia afrelii sp. nov. associated with late cutaneous manifestations of Lyme borreliosis. Scand. J. Infect. Dis., in press.. Picken, R. N. 99. Polymerase chain reaction primers and probes derived from flagellin gene sequences for specific detection of the agents of Lyme disease and North American relapsing fever. J. Clin. Microbiol. 30:99-4.. Postic, D., C. Edlinger, C. Richaud, F. Grimont, Y. Dufresne, P. Perolat, G. Baranton, and P. A. D. Grimont. 990. Two genomic species in Borrelia burgdorferi. Res. Microbiol. 4:465-475. 3. Rosa, P., D. Hogan, and T. G. Schwan. 99. Polymerase chain reaction identify two distinct classes of Borrelia burgdorferi. J. Clin. Microbiol. 9:54-53. 4. Saint Girons, I., and B. E. Davidson. 99. Genome organization of Borrelia burgdorfen in Lyme disease: molecular and immunologic approaches. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 5. Schwartz, J. J., A. Gazumyan, and I. Schwartz. 99. rrna gene organization in the Lyme disease spirochete, Borrelia burgdorfen. J. Bacteriol. 74:3757-3765. 6. Southern, E. M. 975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-57. 7. Stiilhammar-Carlemm, M., E. Jenny, L. Gern, A. Aeschlimann, and J. Meyer. 990. Plasmid anysis and restriction fragment length polymorphisms of chromosom DNA low a distinction between Borrelia burgdorferi isolates. Zentrbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. Orig. 74:8-39. 8. Wlich, R., U. E. Schaible, Y. Lobet, S. E. Moter, M. D. Kramer, and M. M. Simon. 99. Evuation of genetic divergence among Borrelia burgdorferi isolates by use of ospa, fla, HSP60, and HSP70 gene probes. Infect. Immun. 60:4856-4866. 9. Wayne, L. G., D. J. Brenner, R R. Colwell, P. A. D. Grimont, 0. Kandler, M. I. Krichevsky, L. H. Moore, W. E. C. Moore, R. G. E. Murray, E. Stackebrandt, M. P. Starr, and H. G. Truper. 987. Report of the ad hoc committee on reconciliation of approaches to bacteri systematics. Int. J. Syst. Bacteriol. 37:463-464. 30. Welsh, J., Pretzman, D. Postic, I. Saint Girons, G. Baranton, and M. McClelland. 99. Genomic fingerprinting by arbitrarily primed polymerase chain reaction resolves Borrelia burgdorferi into three distinct phyletic groups. Int. J. Syst. Bacteriol. 4:370-377. 3. Wilske, B., V. Preac-Mursic, U. B. Gobel, B. Graf, S. Jauris, E. Soutschek, E. Schwab, and G. Zumstein. 993. An OspA serotyping system for Borrelia burgdorferi based on reactivity with monoclon antibodies and OspA sequence anysis. J. Clin. Microbiol. 3:340-350.